[
https://issues.apache.org/jira/browse/GROOVY-8346?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17050561#comment-17050561
]
Brian Ray commented on GROOVY-8346:
-----------------------------------
[~paulk] would you suggest a distinct issue for this use case?
{code:groovy}
import spock.lang.*
class FinalVariableAnalysisSpec extends Specification {
@Issue( [ "GROOVY-8346", "spockframework/spock#1017" ] )
def "final String variable split across Spock blocks"() {
when: "we define a finalized variable in this block"
final foo = "bar"
//Oddly enough if this block is commented and the downstream and: block
//becomes a then: block, then the FVA problem goes away. And I
//think this has to be a thrown(E) or notThrown(E) flavor of assertion. A
//simple assertion a la 1 == 1 will not trigger the FVA problem. Nor will
//another "wrapper" variant like verifyAll { 1 == 1 }.
then: "no exception"
notThrown(IllegalArgumentException)
and: "we should be able to validate it in a later block"
foo == "bar"
}
}
//[...]FinalVariableAnalysisSpec.groovy: 20: The variable [foo] may be
uninitialized
//. At [20:5] @ line 20, column 5.
// foo == "bar"
//
//With:
// groovy:2.5.9:indy
// spock-core:1.3-groovy-2.5
{code}
I believe under the covers Spock is ASTing verification blocks into
{{try-catch}} blocks, and know that there are one or two issues on this JIRA to
do with FVA and {{try-catch}} (GROOVY-8801 e.g.). So I could tag this on to one
of those.
Relatedly I suspect the Spock project would argue against mitigating this in
their framework given
[spock#1017|https://github.com/spockframework/spock/issues/1017]. But I could
try there.
Last as much as I'm a fan of immutability, I wouldn't mind sweeping all my
tests and switching these to {{def}}. Immutability in test source isn't quite
as important as in application sources.
> The variable [XXX] may be uninitialized
> ----------------------------------------
>
> Key: GROOVY-8346
> URL: https://issues.apache.org/jira/browse/GROOVY-8346
> Project: Groovy
> Issue Type: Bug
> Affects Versions: 2.5.0-beta-1, 2.5.0-beta-2
> Reporter: paolo di tommaso
> Priority: Major
>
> I'm getting several errors when compiling Groovy 2.5.0-beta-2 (and beta-1)
> related to final variable initialisations.
> For example:
> {code}
> [:compileTestGroovy] WorkflowMetadataTest.groovy: 43: The variable [begin]
> may be uninitialized
> [:compileTestGroovy] . At [43:15] @ line 43, column 15.
> [:compileTestGroovy] final begin = new Date()
> [:compileTestGroovy] ^
> {code}
> Related to the following line
> {code}
> final begin = new Date()
> {code}
> See also
> [here|https://github.com/nextflow-io/nextflow/blob/master/src/test/groovy/nextflow/script/WorkflowMetadataTest.groovy#L43-43],
>
> [here|https://github.com/nextflow-io/nextflow/blob/master/src/test/groovy/nextflow/file/SequentialFileStoreTest.groovy#L40-40]
>
> [here|https://github.com/nextflow-io/nextflow/blob/master/src/test/groovy/nextflow/cloud/CloudConfigTest.groovy#L299-299]
> and
> [here|https://github.com/nextflow-io/nextflow/blob/master/src/test/groovy/nextflow/processor/PublishDirTest.groovy#L96-96]
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
